Project Summary: Enveloped viruses enter cells using their fusion proteins, which merge the virus envelope and the cell membrane. Determining the atomic-resolution structures and structural evolution of viral fusion proteins along the fusion pathway is important for understanding how these proteins cause membrane curvature and dehydration and for developing antiviral drugs to inhibit virus entry. Although crystal structures of the water-soluble ectodomains of several viral fusion proteins have been determined, much less is known about the hydrophobic N-terminal fusion peptide (FP) and the C-terminal transmembrane domain (TMD). In the last funding period, we showed that the FP and TMD of the parainfluenza virus 5 (PIV5) fusion protein, F, adopt membrane-dependent conformations. In phosphatidylethanolamine (PE)-rich membranes, the ?-strand conformation dominates and induces negative Gaussian curvature, which is characteristic of hemifusion intermediates, to the membrane. This suggests that the ?-strand structure is important for the prefusion to hemifusion transition. Building on these results, we will now determine the oligomeric structure of the FP and TMD of PIV5 and investigate the structure of the TMD-including region of the HIV fusion protein, gp41. These experiments will address whether the FP and TMD are trimeric when they are apart from each other, as in the early fusion stages, and whether they form a trimer of hairpins when they are brought into close proximity, as in the post-fusion state. In Aim 1, we will measure the oligomeric structure and intermolecular distances of the PIV5 FP and TMD peptides to probe the early- fusion structures. 19F and 13C CODEX experiments will be performed to measure intermolecular distances for both the ?-helical and ?-sheet conformations of the peptides. These experiments will be done as a function of membrane composition, and lipid mixtures that mimic the cell membrane and virus envelope will be used. In Aim 2, we will investigate late-fusion structures of PIV5 by studying a chimera that links the FP and TMD. Vesicle fusion assays show that the chimera is fusogenic, with lipid mixing activities that are approximately the sum of the activities of the FP and TMD peptides. We will develop improved 13C and 1H spin diffusion techniques to measure inter-domain FP-TMD distances as well as intermolecular homo- oligomeric distances. These distance constraints will test whether a six-helix bundle is formed by the FP and TMD inside the membrane. In Aim 3, we will extend our structural studies to HIV gp41, focusing on the antibody-targeted MPER region and the membrane-bound TMD. We will investigate the conformation and curvature-inducing ability of this domain in lipid membranes, conduct 19F-19F and 13C-19F distance experiments to measure the relative orientation of the MPER and TMD, their depths of insertion in the membrane, and the oligomeric structure of this protein. These experiments are expected to provide fundamental new insights into the protein structural transitions that underlie virus-cell fusion.